使用numpy.random.choice：

import numpy as np

L = ['Yes', 'No', 'Info Missing','Not Provided','In Progress']
df['Vaccinaited_or_not'] = np.random.choice(L, size=len(df))

print (df)
      Name  Age Vaccinaited_or_not
0    Aaron   42       Not Provided
1    Steve   44        In Progress
2  Richard   29                Yes

使用np.random.choice：

choices = ['Yes', 'No', 'Info Missing', 'Not Provided', 'In Progress']
df['Vaccinaited_or_not'] = np.random.choice(choices, len(df))
print(df)

# Output
      Name  Age Vaccinaited_or_not
0    Aaron   42        In Progress
1    Steve   44       Not Provided
2  Richard   29       Info Missing

您还可以提供权重：

choices = ['Yes', 'No', 'Info Missing', 'Not Provided', 'In Progress']
weights = [0.2, 0.2, 0.4, 0.1, 0.1]  # sum=1
df['Vaccinaited_or_not'] = np.random.choice(choices, len(df), p=weights)

最佳实践：使用RandomGenerator

choices = ['Yes', 'No', 'Info Missing', 'Not Provided', 'In Progress']
weights = [0.2, 0.2, 0.4, 0.1, 0.1]  # sum must be 1
rng = np.random.default_rng(seed=2023)  # for reproducible case
df['Vaccinaited_or_not'] = rng.choice(choices, len(df), p=weights)

权重说明：使用权重/概率对于理解数据很重要：

x = pd.Series(rng.choice(choices, 10000, p=weights))
x.value_counts(normalize=True).plot.bar(rot=0, title='10000 samples')

输出：x1c 0d1x